Challenge and Dataset on Large-scale Human-centric Video Analysis in Complex Events (HiEve)


Short Name: DeepSORT (Baseline)
Complete name: DeepSORT (Baseline)
Description:

baseline

Runtime:

none

Submissions:

1

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
27.1159 21.9465 28.5525 8.4995 41.4481 5894 42668 2220 90 3122 70.4732

Detailed
performance:


Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 45.1886 43.1755 48.3563 34.6939 14.2857 1564 4678 297 6 542 71.7001
21 13.3599 9.3599 21.3396 0.0000 56.6667 56 778 35 1 62 68.9425
22 3.3996 -1.1285 8.9868 1.9417 73.7864 610 8585 182 11 232 69.8996
23 39.3176 33.9569 31.2673 21.4286 28.5714 797 1827 97 6 190 70.8187
24 7.8550 -3.7694 8.7948 1.2158 65.3495 602 11409 365 39 418 66.3843
25 41.3702 36.1044 22.4090 5.7471 10.3448 168 3030 362 8 338 75.1891
26 34.6237 32.4737 31.8258 23.5294 29.4118 153 2220 59 2 165 69.8674
27 21.2950 15.7832 18.5883 4.8387 20.9677 729 2929 341 7 442 66.9322
28 24.7927 13.7386 27.1547 10.0000 45.0000 308 1522 75 7 124 66.0812
29 37.3268 35.5226 39.0533 16.0000 28.0000 289 1045 68 1 136 68.8685
30 36.2078 36.2078 26.8447 11.1111 17.7778 39 1159 67 0 64 73.4425
31 55.7623 55.7623 49.0729 42.2222 4.4444 175 646 154 0 175 70.3384
32 36.4461 34.9347 44.1552 20.0000 35.0000 404 2840 118 2 234 69.5501